Employing a Machine Learning Approach in Precision Oncology to Predict PIK3CA Functional Status and Identify Phenocopying Variants of Deleterious PIK3CA Mutations - Senior Thesis

Aug 6, 2026

The proposal outlines a modular machine learning pipeline to detect malfunctioning cancer genes and pathways using patient transcriptome data, aiming to reveal hidden responders in precision oncology. Focused on the PI3K/AKT/mTOR pathway, it uses integrated RNA-seq, copy number, and mutation features to predict PIK3CA functional status through learned gene-specific weights. Classifier scores are then tested on cell line datasets with Buparlisib and Copanlisib pharmacological profiling to link predicted functionality with PI3K inhibitor sensitivity, and to support phenocopying variant biomarker discovery.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages32

Reading Time1 h 21 min

AccessFree to download

ABSTRACT

INTRODUCTION

The PI3K/AKT/mTOR Pathway

PI3K Signaling in Cancer

PI3K Inhibitors

Machine Learning Models of Gene Expression

Logistic Regression in Medical and Clinical Studies

Specific Aims

METHODS

Training Pan-Cancer Classifier

Evaluating Pan-Cancer Classifier

Pan-Cancer Classifier Benchmark Analysis

Cell Line Validation

EXPECTED RESULTS

Evaluating Pan-Cancer Classifier

Pan-Cancer Benchmark Analysis

Cell Line Validation

FUTURE DIRECTIONS

ACKNOWLEDGEMENTS

REFERENCES

Claremont Colleges Scholarship @ Claremont | CMC Senior Theses | CMC Student Scholarship | | --- | --- | | 2024
Employing a Machine Learning Approach in Precision Oncology to Predict PIK3CA Functional Status and Identify Phenocopying Variants of Deleterious PIK3CA Mutations
Javier Castillo
Follow this and additional works at: [https://scholarship.claremont.edu/cmc_theses](https://scholarship.claremont.edu/cmc_theses)
Part of the Disease Modeling Commons, Genetic Processes Commons, Neoplasms Commons, and the Other Analytical, Diagnostic and Therapeutic Techniques and Equipment Commons | | Recommended Citation Castillo, Javier, "Employing a Machine Learning Approach in Precision Oncology to Predict PIK3CA Functional Status and Identify Phenocopying Variants of Deleterious PIK3CA Mutations" (2024) . CMC Senior Theses. 3580. [https://scholarship.claremont.edu/cmc_theses/3580](https://scholarship.claremont.edu/cmc_theses/3580) This Open Access Senior Thesis is brought to you by Scholarship@Claremont. It has been accepted for inclusion in this collection by an authorized administrator. For more information, please contact [scholarship@claremont.edu](scholarship@claremont.edu). Employing a Machine Learning Approach in Precision Oncology to Predict PIK3CA Functional Status and Identify Phenocopying Variants of Deleterious PIK3CA Mutations A Thesis Presented by Javier Castillo To the Keck Science Department of Claremont McKenna, Scripps, and Pitzer Colleges In Partial Fulfillment of The Degree of Bachelor of Arts Senior Thesis in Biology April 22, 2024 TABLE OF CONTENTS ABSTRACT....................................................................................................................................2 INTRODUCTION......................................................................................................................... 3 The PI3K/AKT/mTOR Pathway................................................................................................4 PI3K Signaling in Cancer.......................................................................................................... 6 PI3K Inhibitors...........................................................................................................................8 Machine Learning Models of Gene Expression.........................................................................9 Logistic Regression in Medical and Clinical Studies.............................................................. 11 Specific Aims........................................................................................................................... 12 METHODS................................................................................................................................... 13 Training Pan-Cancer Classifier................................................................................................ 13 Evaluating Pan-Cancer Classifier............................................................................................ 16 Pan-Cancer Classifier Benchmark Analysis............................................................................ 17 Cell Line Validation................................................................................................................. 18 EXPECTED RESULTS............................................................................................................... 19 Evaluating Pan-Cancer Classifier............................................................................................ 19 Pan-Cancer Benchmark Analysis............................................................................................ 22 Cell Line Validation................................................................................................................. 24 FUTURE DIRECTIONS.....................................................................................

FAQs

What problem does the machine learning approach address in precision oncology?

It targets malfunctioning genes and pathways in cancer by leveraging transcriptome information that is underused in precision oncology to help identify hidden responders.

How does the method predict PIK3CA functional status?

It integrates RNA-seq, copy number, and mutation data from tumors and uses learned gene-specific weights to infer the functional status of PIK3CA within the PI3K/AKT/mTOR pathway.

How will classifier predictions be evaluated using drug response data?

The trained classifier is applied to cell line datasets that include pharmacological profiling for PI3K inhibitors Buparlisib and Copanlisib, testing whether classifier scores correlate with sensitivity to these drugs.

Aug 6, 2026

Upload by Olivia Brown